The recent discussion on LKML convinced me that a line discipline
is the correct way to layer a driver over a serial interface.
This means, however, that I'll need a (trivial) userspace daemon
which will hold the serial interface open after pushing the line
discipline onto it.
Which raises the question: Where can I drop the source for that
daemon so that it is
a) maintained together with the line discipline in case a kernel
developer feels the need to change something and
b) picked up by distributions so that they don't end up shipping
the line discipline without the necessary daemon?
Is there an appropriate place in or near the kernel tree for
stuff like that?
Thanks
Tilman
--
Tilman Schmidt E-Mail: [email protected]
Bonn, Germany
Diese Nachricht besteht zu 100% aus wiederverwerteten Bits.
Unge?ffnet mindestens haltbar bis: (siehe R?ckseite)
> On Wed, 17 Jan 2007 17:36:14 +0100 Tilman Schmidt <[email protected]> wrote:
> The recent discussion on LKML convinced me that a line discipline
> is the correct way to layer a driver over a serial interface.
> This means, however, that I'll need a (trivial) userspace daemon
> which will hold the serial interface open after pushing the line
> discipline onto it.
>
> Which raises the question: Where can I drop the source for that
> daemon so that it is
> a) maintained together with the line discipline in case a kernel
> developer feels the need to change something and
> b) picked up by distributions so that they don't end up shipping
> the line discipline without the necessary daemon?
>
> Is there an appropriate place in or near the kernel tree for
> stuff like that?
Nope. The requirement arises regularly.
util-linux would seem to be a suitable place, but apparently there
are, umm, isues surrounding that.
Andrew Morton wrote:
>>
>> Is there an appropriate place in or near the kernel tree for
>> stuff like that?
>
> Nope. The requirement arises regularly.
>
> util-linux would seem to be a suitable place, but apparently there
> are, umm, isues surrounding that.
util-linux-ng is already going, so it's a good place.
-hpa
Hi,
On Wed, 07 Feb 2007 14:05:38 -0800, H. Peter Anvin wrote:
> util-linux-ng is already going, so it's a good place.
Not even a year later, I'd like to take up this proposal. I have:
- gigasetm101d.c (the source program)
- gigasetm101d.8 (a manpage)
- Makefile (trivial)
How should I proceed in order to get this included in util-linux-ng?
Thanks,
Tilman
--
Tilman Schmidt E-Mail: [email protected]
Bonn, Germany
Diese Nachricht besteht zu 100% aus wiederverwerteten Bits.
Unge?ffnet mindestens haltbar bis: (siehe R?ckseite)
On Wed, 16 Jan 2008 00:03:43 +0100 Tilman Schmidt wrote:
> Hi,
>
> On Wed, 07 Feb 2007 14:05:38 -0800, H. Peter Anvin wrote:
> > util-linux-ng is already going, so it's a good place.
>
> Not even a year later, I'd like to take up this proposal. I have:
> - gigasetm101d.c (the source program)
> - gigasetm101d.8 (a manpage)
> - Makefile (trivial)
> How should I proceed in order to get this included in util-linux-ng?
See http://userweb.kernel.org/~kzak/util-linux-ng/
for all util-linux-ng info, including mailing list.
---
~Randy
H. Peter Anvin schrieb:
> Andrew Morton wrote:
>>>
>>> Is there an appropriate place in or near the kernel tree for
>>> stuff like that?
>>
>> Nope. The requirement arises regularly.
>>
>> util-linux would seem to be a suitable place, but apparently there
>> are, umm, isues surrounding that.
>
> util-linux-ng is already going, so it's a good place.
I have asked on the util-linux-ng mailing list but received
a flat refusal by Karel:
> I don't think that util-linux-ng is there for this purpose. It's not
> place for HW specific tools. Sorry.
Back to square one.
--
Tilman Schmidt E-Mail: [email protected]
Bonn, Germany
Diese Nachricht besteht zu 100% aus wiederverwerteten Bits.
Unge?ffnet mindestens haltbar bis: (siehe R?ckseite)